Orpheus.cpp is a high-performance text-to-speech (TTS) solution designed for efficient streaming using WebRTC, specifically integrated with FastRTC. This project delivers fast and responsive voice synthesis capabilities, making it an excellent choice for developers looking to implement speech functionalities in their applications.

Key Features

• Fast and Efficient: Leveraging the Orpheus framework, this library provides rapid TTS capabilities suitable for real-time applications.
• Streaming Support: Supports both synchronous and asynchronous streaming modes for flexible usage.
• Easy Integration: Integration with existing Python projects is straightforward, requiring minimal setup and additional dependencies.

Example Usages

• Synchronous TTS: Generate speech output from text using the following code snippet:

  from orpheus_cpp import OrpheusCpp
  from scipy.io.wavfile import write
  
  orpheus = OrpheusCpp()
  text = "I really hope the project deadline doesn't get moved up again."
  # The output consists of a sample rate and a numpy int16 array.
  sample_rate, samples = orpheus.tts(text, options={"voice_id": "tara"})
  write("output.wav", sample_rate, samples.squeeze())
  

• Asynchronous Streaming: Stream text-to-speech results with a non-blocking approach to enhance performance:

  async for sample_rate, samples in orpheus.stream_tts(text, options={"voice_id": "tara"}):
      write("output.wav", sample_rate, samples.squeeze())
  

Streaming Configuration

Users can adjust the pre_buffer_size parameter to optimize streaming responsiveness. A default pre-buffer of 1.5 seconds enables smooth audio delivery but can be adjusted for faster response times:

async for sample_rate, samples in orpheus.stream_tts(text, options={"voice_id": "tara", "pre_buffer_size": 0.5}):
    write("output.wav", sample_rate, samples.squeeze())

With Orpheus C++, developers can create advanced applications requiring high-quality TTS capabilities efficiently, combining speed and simplicity.